Easy weather wild results

[steve]

I'm worried that cumulus has the same problem with corrupted data as easy weather.

Of course it does. Whatever the software is, if the hardware invents incorrect data, then the software will read that data. Cumulus does have a certain amount of filtering, though, of values which are clearly wrong.

Cumulus will then re-download the data from your station (assuming it is a type which has a data logger), and this may well correct the problem

This assumes that the corruption was transient and didn't make it into the data logger. If it did get logged, then it will just get re-downloaded.

The one thing that puzzles me is why a data corruption is not self correcting? Assuming most transfers are accurate & that the logger is downloaded in total every time

Cumulus doesn't download the entire logger every time, just as much as it needs. My experience is that EW does the same. But I have used EW very little.

you would think that subsequent reads would over write corrupted data with good stuff? It never does, I always have to clear the pc data in easy weather then let easy weather get it all again from the logger.

If the bad data has gone into the logger, then rereading it from the logger will recreate the bad data.

Basically, the station (apparently) has no error-checking or correction in its data transmissions, and is susceptible to interference. There appear to be a number of ways that bad data can be introduced.

[KJM]

Steve is the 'certain amount of filtering' you refer to the EWTempdiff for temperatures in the .ini or is there non user adjustable filtering for other parameters built into cumulus as well?

[steve]

There's non-adjustable filtering for other parameters. For example:

RH > 100%
Temp < -50C or > +70C
Pressure < 900mb or > 1200mb
Wind > 60 m/s

At some point I'll probably make these configurable.

[KJM]

Seems like a lot of work for such a flawed device as these Chinese generics seem to be. If I sourced a product like that my customers would run me out of town. So do other stations have this data integrity problem or is most of your filtering effort directed at them exclusively?
Trawling the forum I take your point about different types of data problem in terms of if it's written to the logger or not. My issue with rainfall is corruption of rainfall data in the historical data record easy weather keeps on the pc. Suddenly rainfall will just spontaneously appear in the middle of it. If I clear the data folder, forcing easy weather to reload then the data is good again. I'm starting to think you're right, that easy weather too only downloads new data since the last poll. That would explain why my rain fall glitch that suddenly appears two weeks back in the record does not self repair. If that's the case my particular headache with rainfall is more of a pc database management issue by easy weather.
You've cheered me up greatly Steve. It could well be that the cumulus database might do much better. One question though. For easy weather to corrupt it's data base midway it must write bad data over good. How does cumulus work? Does it extend the database by adding new values after each poll of the logger or could it too conceivably go go back & write over some older records? Just curious?

[steve]

So do other stations have this data integrity problem or is most of your filtering effort directed at them exclusively?

The checks are mainly for Fine Offset and La Crosse stations. Oregon Scientific don't seem to be a problem, and Davis definitely aren't a problem - an occasional glitch perhaps.

How does cumulus work? Does it extend the database by adding new values after each poll of the logger or could it too conceivably go go back & write over some older records? Just curious?

Cumulus never rewrites old data, it always appends.

[KJM]

Neat! I'll see how Cumulus goes with rainfall, if the $100 special gets inventive again I'll try a Davis 6250M, they're down to $US262 online lately. When you say 'occasional glitch' Steve, how infrequent are we talkin' exactly?

[steve]

When you say 'occasional glitch' Steve, how infrequent are we talkin' exactly?

I've had my Davis for over three years, and I think I had one odd unexplained reading once that I had to correct. Apart from that, I did have to replace the temperature sensor when it started giving completely wrong readings, and I had to replace the sensor at the school here at about the same time. I know of others who have had to do the same; possibly Davis had a bad batch of sensors.